create-skill
Pass
Audited by Gen Agent Trust Hub on Mar 6, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is purely instructional and structural, providing a framework for how the agent should help users organize and document their own custom skills. It does not contain any executable logic, prompt injection attempts, or malicious instructions.
- [COMMAND_EXECUTION]: The skill documents the pattern of using local utility scripts (e.g.,
scripts/validate.py) for automated workflows. These references are provided as illustrative examples for users to implement in their own projects and do not involve the execution of untrusted or arbitrary commands by the skill itself. - [EXTERNAL_DOWNLOADS]: The skill mentions standard and well-known software libraries (e.g.,
pdfplumber,pypdf,docx-js) as technical examples of tools users might integrate into their custom skills. No automated network downloads or remote script executions from untrusted sources are present.
Audit Metadata